Neonatal abstinence syndrome

Many illicit substances used by mothers in pregnancy cross the placenta and can cause physical dependency in the fetus. At birth, division of the umbilical cord leads to abrupt cessation of the supply of the illicit drug to the infant, which may result in neonatal abstinence syndrome (NAS). Signs of NAS include irritability, jitteriness, hypertonia, poor feeding, diarrhoea, sweating, skin excoriation and, in extreme cases, convulsions.1 6 7 Although the use of methadone in pregnancy reduces some of the harmful effects of illicit opiate on mother and infant, approximately 40–60% of infants who have been exposed to methadone in utero develop signs of NAS.8 9 10 11 12 13 14 Prescribed maternal methadone dose has not been found consistently to correlate with the development of NAS8 9 10 11 although a recent large retrospective audit of 450 singleton infants born to mothers prescribed methadone in pregnancy reported a strong correlation between prescribed maternal methadone dose and the risk of the infant developing NAS, even when corrected for additional illicit drug use.11 Use of illicit drugs in addition to methadone is associated with increased requirement for treatment of NAS and duration of infant hospital stay,12 and heavy cigarette smoking concurrent with methadone use is associated with more severe symptoms of NAS.13 Breast-feeding reduces the likelihood of requiring treatment for NAS, although the mechanism of this is unclear, as transfer of methadone into maternal milk is minimal.11 16 Development of NAS in infants born to methadone-prescribed mothers may also be influenced by interindividual variation in metabolism of methadone by mother or baby.14 15

As it is impossible to predict the likelihood of an individual baby being affected by NAS, management is usually expectant with many units recommending a prolonged postnatal stay to observe for signs of NAS.11 12 16 Infants who require treatment may need to remain in hospital for many weeks, which has adverse implications for mother and infant bonding, infant development and the cost of neonatal and obstetric services.1 7 11

Treatment options for NAS secondary to maternal opioid use include opiates or opioids, sedatives (phenobarbital or diazepam) and supportive treatments (swaddling, pacifiers, massage).16 17 18 19 20 Cochrane meta-analysis of published studies found a reduction in treatment failure with opiate use compared with the other interventions.18 Following polydrug misuse, a combination of opiate and phenobarbital may reduce the severity of withdrawal and duration of hospitalisation as well as improving neurobehavioural scores.11 17 20

Neurodevelopmental outcome

A recent review article suggests that maternal methadone use is not associated with adverse postnatal development.21 However, there are numerous data linking maternal opiate misuse with developmental delay at ages ranging from 6 months to 12 years.22 23 24 25 26 27 28 29 30 31 32 Various scales of child development have demonstrated motor development delay22 23 24 and low mental development index or low IQ22 23 24 25 26 27 in children born to mothers who have misused opiates in pregnancy. Behavioural problems are also common in children who have been exposed to drug misuse in utero and include attention-deficit/hyperactivity disorder, aggression, poor concentration and social inhibition.25 28 29 30 31 32 Other neurological abnormalities reported include cerebral palsy30 and disorders of muscle tone and posture.23 24 27 Small head circumference is commonly found at birth, usually associated with low birth weight.23 26 27 30

Poor social circumstances and dysfunctional care givers in part account for some of these findings. Ornoy et al attempted to correct for confounding factors by examining the differences in outcome between children raised by their biological parents and those in the care of adoptive parents.30 They found that children in adoptive homes had better development scores than those living with their biological parents, suggesting that the environment plays a significant role in outcome. Children in adoptive homes, however, still had significantly lower scores on their performance scales compared with non-drug-exposed controls.30 Infants born to drug misusing mothers and raised in foster care performed as well as those living with their biological parents but both groups had significantly lower development scores than controls.24 High loss to follow-up is a feature of many of these studies and a reflection of the social disruption associated with the drug culture. Nevertheless it is not unreasonable to speculate that infants whose family life is so chaotic that they are untraceable at follow-up are unlikely to perform better than those infants for whom data are obtainable.22

Visual abnormalities

Prenatal exposure to various harmful substances can have adverse effects on infant visual development. Ophthalmic anomalies following cocaine exposure in utero include strabismus, nystagmus, hypoplastic optic discs, delayed visual maturation and prolonged eyelid oedema.33 34 35 Retinal vascular changes caused by the vasoconstrictive effects of cocaine on placental blood vessels may account for some of these findings.

Ocular abnormalities are also seen in infants with fetal alcohol syndrome, and include short horizontal palpebral fissures, epicanthus, telecanthus, microphthalmia, refractive errors, strabismus and retinal vessel tortuosity.36 37 38 39 Rats exposed to alcohol in utero have hypoplasia of the optic nerve with ultrastructural damage to the macroglial cells and myelin sheaths, and up to half of infants born to alcoholic mothers have optic nerve hypoplasia.39 There are fewer data regarding visual outcome in infants exposed to opiates and/or benzodiazepines in utero. In 49 infants born to opiate-dependent mothers, the rate of strabismus was at least 10 times that of the general population,40 and a recent Scottish study found that 26% of infants born to opiate-using mothers failed vision screening tests on at least one occasion within the first 6 months of life.6 Forty-two per cent of the latter group referred for formal ophthalmology assessment had confirmed abnormalities including nystagmus and squint.6 The presence of nystagmus in five children born to drug-addicted mothers was reported first in 2003.41 Three children presented with congenital horizontal pendular nystagmus and two children with a transient horizontal nystagmus in association with NAS. More recently horizontal nystagmus, strabismus and delayed visual development were described in 14 infants exposed to methadone and/or benzodiazepine in utero.42

In summary, maternal drug misuse frequently results in NAS and can have long-term adverse effects on infant neurological and visual development. The independent effects of maternal drug misuse, NAS and the pharmacological treatment of these conditions, on the developing infant brain and visual pathways are not yet clear. There is a need for greater understanding of the effects of maternal drug misuse on the developing visual system, and this requires measurement of visual function, which can be very challenging in infancy and particularly in the newborn period. Visual evoked potentials (VEPs) can be used to determine the integrity of the visual system and predict both visual and neurological outcome in infants. We discuss the application of the VEP in infancy and summarise current knowledge with regard to the effects of illicit drug use upon the VEP.

Visual evoked potentials

The VEP is an electrical signal generated in the visual cortex of the brain in response to a visual stimulus.43 44 A normal VEP depends upon an intact visual pathway from the retina via the optic nerves and optic chiasm to the lateral geniculate ganglia and visual cortex of the brain. VEPs therefore reflect visual system and cortical integrity, and are a useful measure of visual development. VEPs can be used to detect, quantify and monitor abnormalities of the visual system.44

Several stimuli can be used to elicit VEPs in infants; those most frequently used are a flashing or flickering light source and a black and white checkerboard pattern.43 44 The flash VEP is most commonly measured in the newborn, as it does not require visual fixation.45 The flash VEP is first detectable at around 24 weeks’ gestation, and the waveform matures progressively thereafter to term44 45 46 (fig 1A). Recording VEPs to a pattern stimulus produces less intrasubject variability and allows assessment of visual acuity48 (fig 1B), but requires the infant to fixate on a visual display and consequently is of limited applicability in the newborn period.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

(A) Flash visual evoked potentials (VEP). A flash light stimulus produces a waveform with positive components at approximately 100 ms (P1) and 200 ms (P2) and a negative component at approximately 300 ms (N3). (B) Pattern VEP. Black and white reversing checkerboards produce a negative–positive–negative waveform, conventionally labelled N75-P100-N145: however, in infants, it is normal for the P100 peak to be as late as 200 ms, depending on their age and the size of the checkerboard. (C) Flicker VEP. A flickering light stimulus produces a continuous oscillating waveform (1). Mathematical analysis shows a significant response at 5 Hz, demonstrating the brain’s response to a 5 Hz flickering light (2).

An alternative and novel electrophysiology technique applicable in the newborn period is the flicker VEP, whereby a flickering light source elicits VEPs which overlap in time to produce a continuous oscillating waveform. Using a mathematical technique to interpret the data, the brain’s response to the flickering light can be determined.47 Like the flash VEP, the flicker VEP does not require visual fixation, and it may provide an alternative practical indicator of visual pathway integrity and maturation47 (fig 1C).

Application of visual evoked potentials in infants

In clinical practice, the VEP is used to test whether the visual system is intact, and can detect optic nerve abnormalities (such as hypoplasia) and/or posterior visual pathway abnormalities.44

The VEP can also be used to predict adverse neurological outcome in preterm infants as well as in term infants with birth asphyxia49 50 51 52 53 54 55 56 57 58 59 60 (table 1). In preterm infants with intraventricular haemorrhage and periventricular leucomalacia, the flash VEP predicts likelihood of death and/or cerebral palsy. Poor prognostic factors for outcome include an absent VEP at any age, delayed N3 latency before expected date of delivery and an absent P2 component at term.52 53 54 In term infants with birth asphyxia and resultant hypoxic–ischaemic encephalopathy, death or severe neurological impairment can be predicted by an absent flash VEP at any time or if abnormalities of the waveform persist beyond the fourth day of life.55 56 57 58 The pattern reversal VEP also correlates with developmental outcome in preterm infants59 and with cerebral damage as assessed by MRI in term infants.60 In infants with acute cortical blindness or birth asphyxia, the VEP is highly sensitive in predicting long-term visual outcome.50 51

Visual evoked potentials and drug misuse

In adults on maintenance methadone therapy, the pattern VEP demonstrates a delay in N75 and P100 components compared with normal controls.61 This is postulated to be due to an adverse effect of opioid exposure on neural transmission within primary visual areas of the brain. Previous cocaine use is also associated with a significantly delayed P100 latency in the pattern VEP,62 which could be explained by the vasoconstrictive effect of cocaine on the retinal and occipital vasculature.62 Similarly, chronic alcoholism has an adverse effect on the pattern VEP with reported abnormalities including a delayed P100 as well as an abnormal waveform.63 64

Rat pups born to methadone-exposed mothers demonstrated abnormal flash and flicker VEPs during the abstinence syndrome, although all VEPs had normalised by 21 days of life.65

We have recently compared flash VEPs recorded in the first few days after birth in infants exposed to methadone in utero with those of control infants.66 Two-thirds of drug-exposed infants underwent repeat studies after 1 week. Only 76% of the methadone-exposed infants had a detectable flash VEP in the first few days of life, compared with 100% of control infants. VEPs in methadone-exposed infants were of immature waveform, and of significantly lower amplitude than in controls. After 1 week, VEPs in the methadone-exposed infants had matured but remained of smaller amplitude and were still more likely to be undetectable compared with VEPs in newborn non-drug exposed controls. The majority of infants in this study were exposed to illicit drugs in addition to methadone, and in the absence of longer-term follow-up, it is unclear whether changes in neonatal VEPs are due to transient effects of circulating methadone and/or other substances of misuse, or whether they have long-term implications for visual function. The study findings do, however, accord with the delayed visual maturation previously reported in maternal drug-exposed infants.42 The effects of drugs of misuse on the visual system may therefore be threefold. There appears to be an immediate pharmacological effect secondary to circulating drug as shown in adult and animal studies and consistent with our pilot study of newborn infants. There may also be an effect upon the visual system due to NAS itself as suggested by Gaillard and Borruat.41 Finally, there is a longer-term adverse effect on visual development due to a teratogenic effect of drug exposure in utero.6 40 42

Future

Visual and developmental problems are common in infants exposed to drug misuse in utero. Visual and neurodevelopmental abnormalities can be predicted by abnormal VEPs in infancy, and drug misuse is associated with an alteration of the VEP in adults and animal models as well as in the newborn infant.

The role of the VEP in the early detection of NAS and the visual and neurodevelopmental abnormalities secondary to maternal drug misuse warrants further investigation. Long-term follow-up of a large number of exposed infants, along with comprehensive toxicology, is required to determine the relationship between the VEP and prenatal exposure to opiates and other illicit drugs. If the VEP were found to correlate with NAS or longer-term adverse outcome, this could improve hospital management of NAS and/or trigger earlier referral to appropriate specialists. This would have the potential to improve visual and developmental outcomes secondary to early intervention in this vulnerable group of infants.

Infants born to drug-misusing mothers represent a highly vulnerable group who deserve greater attention with regard to both prediction and treatment of NAS and monitoring of visual and neurological development—we must take care not to lose sight of these babies.